Log In Sign Up

Hierarchical Inducing Point Gaussian Process for Inter-domain Observations

by   Luhuan Wu, et al.

We examine the general problem of inter-domain Gaussian Processes (GPs): problems where the GP realization and the noisy observations of that realization lie on different domains. When the mapping between those domains is linear, such as integration or differentiation, inference is still closed form. However, many of the scaling and approximation techniques that our community has developed do not apply to this setting. In this work, we introduce the hierarchical inducing point GP (HIP-GP), a scalable inter-domain GP inference method that enables us to improve the approximation accuracy by increasing the number of inducing points to the millions. HIP-GP, which relies on inducing points with grid structure and a stationary kernel assumption, is suitable for low-dimensional problems. In developing HIP-GP, we introduce (1) a fast whitening strategy, and (2) a novel preconditioner for conjugate gradients which can be helpful in general GP settings.


page 8

page 15

page 16


Kernel Interpolation for Scalable Structured Gaussian Processes (KISS-GP)

We introduce a new structured kernel interpolation (SKI) framework, whic...

Scalable Gaussian Processes with Grid-Structured Eigenfunctions (GP-GRIEF)

We introduce a kernel approximation strategy that enables computation of...

Thoughts on Massively Scalable Gaussian Processes

We introduce a framework and early results for massively scalable Gaussi...

Toward a diagnostic toolkit for linear models with Gaussian-process distributed random effects

Gaussian processes (GPs) are widely used as distributions of random effe...

Efficient Gaussian Process Classification Using Polya-Gamma Data Augmentation

We propose an efficient stochastic variational approach to GP classifica...

Gaussian Processes on Hypergraphs

We derive a Matern Gaussian process (GP) on the vertices of a hypergraph...